==3216== Cachegrind, a cache and branch-prediction profiler
==3216== Copyright (C) 2002-2017, and GNU GPL'd, by Nicholas Nethercote et al.
==3216== Using Valgrind-3.19.0 and LibVEX; rerun with -h for copyright info
==3216== Command: _bin/cxx-opt-sh/mycpp-nosouffle/osh -n --ast-format none benchmarks/testdata/ltmain.sh
==3216== Parent PID: 3209
==3216== 
--3216-- warning: L3 cache found, using its data for the LL simulation.
==3216== brk segment overflow in thread #1: can't grow to 0x4840000
==3216== (see section Limitations in user manual)
==3216== NOTE: further instances of this message will not be shown
==3216== 
==3216== I   refs:      112,486,448
==3216== I1  misses:      1,175,294
==3216== LLi misses:          8,841
==3216== I1  miss rate:        1.04%
==3216== LLi miss rate:        0.01%
==3216== 
==3216== D   refs:       53,105,216  (27,609,873 rd   + 25,495,343 wr)
==3216== D1  misses:        657,330  (   373,802 rd   +    283,528 wr)
==3216== LLd misses:        166,094  (    12,797 rd   +    153,297 wr)
==3216== D1  miss rate:         1.2% (       1.4%     +        1.1%  )
==3216== LLd miss rate:         0.3% (       0.0%     +        0.6%  )
==3216== 
==3216== LL refs:         1,832,624  ( 1,549,096 rd   +    283,528 wr)
==3216== LL misses:         174,935  (    21,638 rd   +    153,297 wr)
==3216== LL miss rate:          0.1% (       0.0%     +        0.6%  )